We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Adsorption of Copper(II) Ion from Aqueous Solution Using Biochar Derived from Rambutan (Nepheliumlappaceum) Peel: Feedforward Neural Network Modelling Study.
- Authors
Selvanathan, Manimala; Yann, Khoo; Chung, Chang; Selvarajoo, Anurita; Arumugasamy, Senthil; Sethu, Vasanthi
- Abstract
Biochars, derived from rambutan (Nepheliumlappaceum) peel through slow pyrolysis, were characterised and investigated as potential adsorbent for the removal of copper ion, Cu(II) from aqueous solution. Characteristics of five biochars of rambutan peel with different pyrolytic temperatures ranging from 300 to 700 °C (B300, B400, B500, B600, B700) were studied, and adsorption abilities of respective biochars were evaluated. Adsorption experiments were carried out by varying adsorbent dosage (0.2, 0.4, 0.8, 1.0, 2.0, and 4.0 g/L) and initial copper ion, Cu(II) concentrations (50 and 100 mg/L) to determine the optimum pyrolytic temperature of biochar with high adsorption affinity. The adsorption kinetics were best described by the pseudo-second order model for all the tested biochars, while the adsorption equilibrium best fitted by Langmuir isotherm. The overall results showed that biochar derived at 600 °C can be used as an effective adsorbent for removal of Cu(II) from aqueous solutions. Furthermore, feedforward artificial neural network (FFBP) modelling was performed to compare the simulated results with experimental output data of Thermogravimetric analysis (TGA) and atomic absorption spectroscopy (AAS) analysis which were trained using Levenberg-Marquardt (LM) backpropagation algorithm. The FFBP structure for pyrolysis process comprised of TGA temperature as input and biomass final weight as output. The adsorption modelling was simulated using adsorption time, temperature, biochar dosage and initial Cu(II) concentration as input data, while final Cu(II) concentration was used as output data to the network. Finally, modelling structure of 1-9-1 and 4-8-1 gave best performance with regression, R value of 0.9999 and 0.9547 for TGA and AAS analysis, respectively.
- Subjects
COPPER absorption &; adsorption; COPPER ions; BIOCHAR; RAMBUTAN; FEEDFORWARD neural networks
- Publication
Water, Air & Soil Pollution, 2017, Vol 228, Issue 8, p1
- ISSN
0049-6979
- Publication type
Article
- DOI
10.1007/s11270-017-3472-8